Cox model and decision trees: an application to breast cancer data

Pereira, Leonardo Costa; Silva, Sóstenes Jerônimo da; Fidelis, Cleanderson Romualdo; Brito, Alisson de Lima; Júnior, Sílvio Fernando Alves Xavier; Andrade, Lorena Sofia dos Santos; Oliveira, Milena Edite Casé de; Oliveira, Tiago Almeida de

doi:10.26633/rpsp.2022.17

Cited by 5 publications

(7 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We have considered conditional inference trees for survival analysis with censored data, which do not assume the need for proportional hazards and have the flexibility to model curves with different shapes for identified groups of subjects. Such trees estimate a regression relationship by recursive binary partitioning in a conditional inference structure, which ensures adequate tree growth without the need for further cross-validation [17]. The algorithm works in three steps.…”

Section: Discussionmentioning

confidence: 99%

A Novel Decision Tree Model for Predicting the Cancer-Specific Survival of Patients with Bladder Cancer Treated with Radical Cystectomy

Sarrio-Sanz,

Martinez-Cayuelas,

Beltran-Perez

et al. 2024

JCM

View full text Add to dashboard Cite

Background/Objectives: The aim was to develop a decision tree and a new prognostic tool to predict cancer-specific survival in patients with urothelial bladder cancer treated with radical cystectomy. Methods: A total of 11,834 patients with bladder cancer treated with radical cystectomy between 2004 and 2019 from the SEER database were randomly split into the derivation (n = 7889) and validation cohorts (n = 3945). Survival curves were estimated using conditional decision tree analysis. We used Multiple Imputation by Chained Equations for the treatment of missing values and the pec package to compare the predictive performance. We extracted data from our model following CHARMS and assessed the risk of bias and applicability with PROBAST. Results: A total of 4824 (41%) patients died during the follow-up period due to bladder cancer. A decision tree was made and 12 groups were obtained. Patients with a higher AJCC stage and older age have a worse prognosis. The risk groups were summarized into high, intermediate and low risk. The integrated Brier scores between 0 and 191 months for the bootstrap estimates of the prediction error are the lowest for our conditional survival tree (0.189). The model showed a low risk of bias and low concern about applicability. The results must be externally validated. Conclusions: Decision tree analysis is a useful tool with significant discrimination. With this tool, we were able to stratify patients into 12 subgroups and 3 risk groups with a low risk of bias and low concern about applicability.

show abstract

Section: Discussionmentioning

confidence: 99%

A Novel Decision Tree Model for Predicting the Cancer-Specific Survival of Patients with Bladder Cancer Treated with Radical Cystectomy

Sarrio-Sanz,

Martinez-Cayuelas,

Beltran-Perez

et al. 2024

JCM

View full text Add to dashboard Cite

show abstract

“…This method is widely used to model survival after diagnosis depending on various risk factors (hazards). For instance in [15][16][17][18], the Cox proportionate hazards method is used for survival analysis. An interesting approach is given by Putter et al [19], where the authors suggest the way how to deal with time dependent variables when using the Cox model for survival analysis among lung cancer patients.…”

Section: Of 13mentioning

confidence: 99%

Actuarial Analysis of Survival after Breast Cancer Diagnosis among Lithuanian Females

Levickytė,

Skucaitė,

Šiaulys

et al. 2024

Preprint

View full text Add to dashboard Cite

Breast cancer is the most common cause of mortality due to cancer for woman both in Lithuania and worldwide. Chances of survival after diagnosis differ significantly depending on the stage of disease at the time of diagnosis and other factors. One way to estimate survival is to construct a Kaplan-Meier estimate for each factor value separately. However, in cases when it is impossible to observe a large number of patients (for example, in case of countries with lower numbers of inhabitants) dividing data into subsets, say, by stage at diagnosis may lead to results where some subsets contain too little data so that results of Kaplan Meier (or any other) method will become statistically incredible. The problem may become even more acute if the researcher would like to use more risk factors, such as stage at diagnosis, sex, place of living, treatment method, etc. Alternatively, Cox models are used to analyse survival data with covariates, and they don’t require dividing the data into subsets according to chosen risks factors (hazards).

show abstract

“…7,11 Evidências sobre análise do AM e seus fatores associados em mães adolescentes são escassas, 8,9,14 principalmente aquelas que abordam a continuidade do AM após a alta hospitalar. 8 A relevância da atual pesquisa, portanto, apresenta-se no fato de abordar um assunto importante e de poucas publicações em âmbito nacional e internacional, 5,14,20…”

Section: Variáveis Independentesunclassified

Prevalência de aleitamento materno e fatores associados entre mães adolescentes de Governador Valadares, Minas Gerais

Izidoro

Chitarra

Silva

et al. 2022

hu rev

View full text Add to dashboard Cite

Introdução: O aleitamento materno (AM) é recomendado pela Organização Mundial da Saúde de forma exclusiva até seis meses de vida e complementado até ≥2 anos. Para as nutrizes, a amamentação oferece benefícios como proteção para diabetes tipo II, retorno mais rápido ao peso pré-gestacional e aumento do espaçamento entre gestações. Entretanto, a prevalência da amamentação no Brasil (2013) foi de apenas 56%, sendo a adolescência fator de risco para a não amamentação e o desmame precoce. Objetivo: Analisar a prevalência de AM aos quatro meses após o parto e seus fatores associados entre mães adolescentes do município de Governador Valadares, MG. Material e Métodos: Estudo transversal, parte de pesquisa intitulada “Consumo alimentar de gestantes adolescentes e retenção de peso pós-parto: um estudo de coorte”. Realizou-se um censo abrangendo todas as puérperas adolescentes (idade <20 anos) residentes no município que tiveram seu parto nas três maternidades locais entre outubro de 2018 e outubro de 2019. A coleta de dados ocorreu por questionário nas primeiras 48 horas pós-parto e no 4º mês pós-parto. Os dados foram analisados no software Stata®16.1. Resultados: Foram entrevistadas 367 mães (taxa de resposta 98,6%) com idade média de 17,6 anos (±1,57). Quatro meses após o parto realizou-se visita domiciliar, compreendendo 317 mães. Destas, 75,4% mantiveram a amamentação e somente 25,9% ofereciam exclusivamente leite materno. Conclusão: Verifica-se que, apesar da elevada intenção de amamentar, há baixa prevalência de AM exclusivo ao 4º mês pós-parto. Menor escolaridade, tabagismo, menor idade materna e trabalhar fora de casa apresentaram-se como fatores de risco para menor tempo de manutenção do AM. Deve-se considerar que a lactação é envolta por grande carga emocional e, na adolescência, somam-se outros fatores psicológicos, fisiológicos e inexperiência para lidar com a maternidade, sendo necessária uma forte rede de apoio profissional durante os períodos pré-natal, parto e pós-parto.

show abstract

Cox model and decision trees: an application to breast cancer data

Cited by 5 publications

References 27 publications

A Novel Decision Tree Model for Predicting the Cancer-Specific Survival of Patients with Bladder Cancer Treated with Radical Cystectomy

A Novel Decision Tree Model for Predicting the Cancer-Specific Survival of Patients with Bladder Cancer Treated with Radical Cystectomy

Actuarial Analysis of Survival after Breast Cancer Diagnosis among Lithuanian Females

Prevalência de aleitamento materno e fatores associados entre mães adolescentes de Governador Valadares, Minas Gerais

Contact Info

Product

Resources

About